End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -32,9 +34,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_w_init__mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6258
+- Loss: 0.6237
 - Accuracy: 0.6838
 - F1: 0.8122
 - Combined Score: 0.7480

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.6237123012542725,
+    "eval_runtime": 0.7659,
+    "eval_samples": 408,
+    "eval_samples_per_second": 532.701,
+    "eval_steps_per_second": 5.223,
+    "train_loss": 0.9404267789760312,
+    "train_runtime": 225.9356,
+    "train_samples": 3668,
+    "train_samples_per_second": 811.736,
+    "train_steps_per_second": 6.418
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 0.6237123012542725,
+    "eval_runtime": 0.7659,
+    "eval_samples": 408,
+    "eval_samples_per_second": 532.701,
+    "eval_steps_per_second": 5.223
+}

logs/events.out.tfevents.1685528848.serv-9223.3032138.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95fb39eff698b5ed9335e718e3efe8a30c373c4d983ddabf1f91ec77afd164ec
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.9404267789760312,
+    "train_runtime": 225.9356,
+    "train_samples": 3668,
+    "train_samples_per_second": 811.736,
+    "train_steps_per_second": 6.418
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_metric": 0.6237123012542725,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mrpc/checkpoint-116",
+  "epoch": 9.0,
+  "global_step": 261,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00049,
+      "loss": 3.2785,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6238367557525635,
+      "eval_runtime": 0.7409,
+      "eval_samples_per_second": 550.699,
+      "eval_steps_per_second": 5.399,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 0.7343,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6785768866539001,
+      "eval_runtime": 0.7398,
+      "eval_samples_per_second": 551.485,
+      "eval_steps_per_second": 5.407,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 0.6377,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.62451171875,
+      "eval_runtime": 0.7405,
+      "eval_samples_per_second": 550.974,
+      "eval_steps_per_second": 5.402,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 0.6353,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6237123012542725,
+      "eval_runtime": 0.7399,
+      "eval_samples_per_second": 551.451,
+      "eval_steps_per_second": 5.406,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 0.6344,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6243776679039001,
+      "eval_runtime": 0.7419,
+      "eval_samples_per_second": 549.914,
+      "eval_steps_per_second": 5.391,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 0.6314,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6323529481887817,
+      "eval_runtime": 0.7418,
+      "eval_samples_per_second": 550.048,
+      "eval_steps_per_second": 5.393,
+      "step": 174
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 0.6431,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6401510834693909,
+      "eval_runtime": 0.7429,
+      "eval_samples_per_second": 549.206,
+      "eval_steps_per_second": 5.384,
+      "step": 203
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 0.6347,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6335784196853638,
+      "eval_runtime": 0.7439,
+      "eval_samples_per_second": 548.43,
+      "eval_steps_per_second": 5.377,
+      "step": 232
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00041,
+      "loss": 0.6343,
+      "step": 261
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6257611513137817,
+      "eval_runtime": 0.7412,
+      "eval_samples_per_second": 550.441,
+      "eval_steps_per_second": 5.396,
+      "step": 261
+    },
+    {
+      "epoch": 9.0,
+      "step": 261,
+      "total_flos": 4888970768941056.0,
+      "train_loss": 0.9404267789760312,
+      "train_runtime": 225.9356,
+      "train_samples_per_second": 811.736,
+      "train_steps_per_second": 6.418
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 4888970768941056.0,
+  "trial_name": null,
+  "trial_params": null
+}